[
https://issues.apache.org/jira/browse/HADOOP-13738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15616874#comment-15616874
]
Arpit Agarwal commented on HADOOP-13738:
----------------------------------------
Thanks [~anu]. Yeah we should hold off committing for a few days to let Kihwal
and Xiaoyu comment.
bq. As for the I/O getting stuck in a disk, I am hoping the normal I/O on that
datanode would have caught that issue and errors already propagated. So I am
okay with not solving that in DiskChecker.
Many IO failures in the DataNode trigger DiskChecker. So if the disk is failing
it should eventually affect DiskChecker.
> DiskChecker should perform some disk IO
> ---------------------------------------
>
> Key: HADOOP-13738
> URL: https://issues.apache.org/jira/browse/HADOOP-13738
> Project: Hadoop Common
> Issue Type: Improvement
> Reporter: Arpit Agarwal
> Assignee: Arpit Agarwal
> Attachments: HADOOP-13738.01.patch, HADOOP-13738.02.patch,
> HADOOP-13738.03.patch, HADOOP-13738.04.patch
>
>
> DiskChecker can fail to detect total disk/controller failures indefinitely.
> We have seen this in real clusters. DiskChecker performs simple
> permissions-based checks on directories which do not guarantee that any disk
> IO will be attempted.
> A simple improvement is to write some data and flush it to the disk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
